Between Tool and Trouble: Student Attitudes Toward AI in Programming Education

TL;DR

This paper investigates how AI code assistants influence novice programmers' experiences, perceptions, and challenges during programming tasks, revealing benefits in understanding and confidence but also issues with overreliance and knowledge transfer.

Contribution

It provides empirical insights into student attitudes and challenges with AI tools in programming education, emphasizing the need for effective pedagogical integration.

Findings

AI tools aid understanding and confidence during initial coding

Students struggle to transfer skills without AI support

Overreliance on AI may hinder conceptual learning

Abstract

This study examines how AI code assistants shape novice programmers experiences during a two-part exam in an introductory programming course. In the first part, students completed a programming task with access to AI support; in the second, they extended their solutions without AI. We collected Likert-scale and open-ended responses from 20 students to evaluate their perceptions and challenges. Findings suggest that AI tools were perceived as helpful for understanding code and increasing confidence, particularly during initial development. However, students reported difficulties transferring knowledge to unaided tasks, revealing possible overreliance and gaps in conceptual understanding. These insights highlight the need for pedagogical strategies that integrate AI meaningfully while reinforcing foundational programming skills.